Restoration of aquifers Methods for identifying, containing, and reclaiming contaminated groundwater are being developed and refined. But rehabilitation costs could remain high Despite legislation aimed at protecting groundwater supplies, such as the Safe Drinking Water Act (SDWA) and the Resource Conservation and Recovery Act (RCRA) and related regulations, aquifers do become contaminated. When this occurs, there are three basic courses of action: Forbid use of the aquifer and obtain alternative water supplies, attempt to rehabilitate it, or continue to use the aquifer, but treat the water to remove the contaminants. About 50% of the total of 3 X 10” acre-ft (1 acre-ft = 325 900 gal) of groundwater estimated to be available in the US. is designated for drinking water. Presently, less than 1% of that total, or 3 X IO9 acre-ft, is believed to be contaminated. A national goal is to halt further groundwater contamination and, if and when feasible, to rehabilitate contaminated water sup plies. Abandoning a contaminated aquifer and procuring water supplies elsewhere might seem to be a more expedient way to solve the problem of contaminated groundwater. But this has disadvantages. Alternative supplies might be neither as cheap nor as abundant as one would hope. Moreover, if groundwater contamination ever threatened to become a national phenomenon, and more well systems were abandoned, alternative water supplies could become scarce, poorer in quality, and expensive. Kirk Brown of Texas A&M University told ES& T that once an aquifer used as a water supply is contaminated, 100% of its available groundwater must be considered contaminated. And rehabilitating groundwater is neither cheap nor easy.
that contamination incidents tend to occur near populated areas that depend upon this water for potable supplies. Significant episodes, some of which forced well abandonment, have occurred in parts of Long Island, New Jersey, Massachusetts, Tennessee, Florida, and several other states. Jay Lehr, executive director of the National Water Well Association (NWWA, Worthington, Ohio), helieves that the pendulum will swing from well abandonment toward aquifer rehabilitation in response to contamination incidents. “People have reached their perceived limit of tolerance to groundwater contamination,” Lehr told ES&T. “That’s not because of health threats; people are not dying. But they want to clean up this water just as they clean their houses or yards when they see a mess there.” He added that in his view, “the U.S. has the technology and know-how to clean up groundwater supplies; aquifer restoration will be a major American industry within the next several years.” Lehr noted that at present, about 75 reclamation or cojtainment projects are under way or planned. Whether Lehr’s optimistic predictions concerning the growth of an aquifer restoration industry will materialize remains to be seen. One thing is certain-such restoration work will require major scientific and technological efforts and outlays of funds. Contamination plumes must be identified, mapped, monitored, and contained. Then a treatment and use or recharge scheme must be designed, installed, and operated. Costs could run into millions or tens of millions of dollars for each aquifer rehabilitation project.
Perceived limit of tolerance Although the total amount of contaminated groundwater nationwide comprises only a small portion of available groundwater, a problem is
Plume identification The first step toward groundwater restoration is knowing the extent, thickness, rate of spread, and contents of a contamination plume. Until re-
0013-936X/83/0916-0347A$01.50/0
@ 1983 American Chemical Soclely
,-
Aquifer restoration: treatment alternatives
tion and choice of treatment alterna. tives. The ldentification step in. dudes: Defining the site‘s environmental setting Surface Subsurface
I
migration pathways Determining contaminant (plume) extent and concentration. Candidate treatment processes
chemical treatment Chemical coagulation pH adjustment Carbon adsorption (liquid and vapor phase) Air skipping Steam stripping Resin adsorption Chemical oxidation Biological treatment Aerobic fixed film Aerobic suspended growth Residuals.treaimnt
Envitm. Sci. Tschnol., VoI. 17, NO.8.1883
341&
cently, about the only way to map the plume was to drill numerous monitoring wells around what is believed to be the plume’s perimeter. Moreover, the plume’s thickness might not be as great as that of the aquifer itself; so if a monitoring well is not screened properly, its intake might miss the contaminant plume altogether since groundwater flow and contaminant transport can be extremely stratified. Conceivably, the zone of contamination could be both overlain and underlain by clean water. That is why the utmost care must be taken, when wells are screened, to account for the three-dimensional aspect of the plume when a monitoring well network is planned. In fact, in a sense, plume identification involvesfour dimensions. The monitoring scheme must consider plume spread in two horizontal and one vertical direction over time. An example of a stratified contamination plume containing volatile organic solvents, in Nashua, N.H., was described a t the Third National Symposium and Exposition on Aqujfer Restoration and Groundwater Monitoring, held at Columbus, Ohio, in late May. Matthew Barvenik, a senior geotechnical engineer with Goldberg-Zoino & Associates, Inc., and a
3 340A
Envimn. Sci. Technoi.. Vol. 17, NO. 8. 1983
principal of Barcad Systems, Inc. (Mass.), spoke of a plume more than 450 m long, up to 33 m deep, covering about 1.2 X 105 m*, and moving at cm/s. It originated about 7 X from the disposal of drums, chemical sludges, demolition debris, and domestic rubbish in an abandoned 24 OOO-mZsand and gravel pit. In addition, more than an estimated 4 X IO6 L of liquid chemical wastes had been discharged next to the pit by “midnight dumpers.” Site cleanup is being carried out under the first cooperative agreement signed pursuant to the federal “Superfund” law. Total construction costs, including change orders, are $2.4 million. Estimates of costs for the completed project come to $10 million. Work on the site has been under way since 1980. To define the plume, more than 60 single-level and multilevel monitoring wells were installed. A properly installed well system must monitor the entire thickness of the aquifer, while taking into account the stratified nature of water movement and contaminant transport. In this particular case, monitoring would have to go as deep as 33 m. Typically, the cost of such wells could range from $400 to $3000 each. Add to these figures the costs associ-
ated with sampling and analysis, which might come to $200-$1500/sample. In estimating costs, one should also include the expenses of hiring a consulting firm, Kirk Brown suggests. While monitoring wells will always have to be used to delineate groundwater contamination, other, less expensive methods might be employed to obtain at least a rough idea of where the wells can be placed most effectively and what substances they should monitor. In many situations, geophysical techniques can perform these tasks and can also provide profiling and area mapping and knowledge of the geological media through which groundwater and its contaminants migrate. Donald Bruehl of Normandeau Associates Inc. (Bedford, N.H.) lists electrical resistivity sounding, seismic refraction profiling, and precision gravity surveys as “methods with which we have had much success in groundwater studies.” Other techniques he mentions include magnetic and electromagnetic methods, earth conductivity, borehole geophysics, and remote sensing. He has used these techniques in groundwater surveys “in geologic conditions typical of the formerly glaciated Northeast.” These conditions usually entail aquifers
consisting of stratified, glacially deposited sand and gravel. These materials are very permeable. Other contamination incidents have involved aquifers of glacial till (mixed clay, sand, gravel, and boulders) and fractured bedrock (permeable through the fractures or cracks in the rock). Resistivity measurements have been used as a preliminary method of defining the horizontal and vertical extent of leachate plumes stemming from suspected hazardous waste drum burial areas in EPA Region 4 (southeastern US.). On the other hand, magnetometry can be useful for detecting contamination “source areas” in the form of buried steel drums. A. Shane Hitchcock of EPA Region 4 (Atlanta, Ga.) explains that where such technologies can be used, they reduce equipment costs, provide data immediately, and, with sufficient background information, can be very reliable. Plume containment The strategy for containing the groundwater contaminants in Nashua, N.H., consisted of installing a I-mthick wall to surround them. The wall is a homogeneous soil-bentonite mixture constructed by a slurry techniaue: the material, keyid into bedrock at a maximum depth of 33 m is allowed to set. The wall is also keyed into a membrane surface cap. After it has set, the wall’s hydraulic conductivity is specified not to exceed IO-’ cm/s-the typical impermeability standard for hazardous waste sites-with leachate as permeant, Barvenik said. Moreover, since seepage under the wall through the rock was anticipated, a leachate collection and treatment plant has also been designed for the project; the design life of the containment system is 30 y. But there are those who look askance at numbers indicating low hydraulic conductivity and permeability. Texas A&M’s Kirk -Brown maintains that under certain conditions, such as exposure to organic wastes, normally impermeable clays undergo structural changes and become more permeable, sometimes by several orders of magnitude (ES&T, September 1982, p. 503A). Perhaps this is one reason why EPA calls for synthetic impermeable materials in liners meant to protect groundwater from hazardous substances. Treatment alternatives Once a contamination plume is identified and mapped, a treatment scheme is needed. Treatment alterna-
Treatment: Mohile unir for grounduurur und leuchorr irrurn,mf i m s PAC’.
tives are considered, and then benchand pilot-scale studies are conducted. Data from these studies help to establish design criteria for various treatment processes and to demonstrate performance capabilities of candidate processes, said John Absalon of Engineering-Science Inc. (Atlanta, Ga.). Treatability studies, in turn, can be used to develop a process aimed at specific problems, such as chlorinated solvents.~hydrocarbons. heavy metals. and other contaminants. The treatment of contaminated aquifer water could be carried out to prevent contamination spread (with purge wells) or to provide process water, or even potable water. Robert OBrien of Calgon Carbon Corporation (Pittsburgh, Pa.) explains that granular activated carbon (GAC) is a frequently adopted treatment for contamination episodes involving organic chemicals. He says that his company installed GAC systems at 31 sites in the U.S. to treat contaminated groundwater flows of 5-2250 gallons per minute (gpm). In some cases, multimedia filtration was done before GAC treatment as a precaution; in others, air stripping was used to reduce the amount of;olatile organics. OBrien savs that for carbon tetrachloride (at four sites), influent concentrationsof 130-10 000pg/Lwere reduced to below 1 pg/L, and 5-16 000pg/Loftrichloroethylene(15 instances) were reduced to less than 1 pg/L. At one site, DDT, present at 4 pg/L, was curtailed to less than 0.05 pg/L. OBrien notes that pre- and post-treatment analyses conformed to applicable EPA-published Drotocols for trace organ& in water.’He estimates that oDeratine costs are $0.22-$2.52/1000 gal-treated, de: pending upon the chemical(s) and their concentrations. ~~
Donald Hager of Rubel and Hager, Inc. (Tucson, A r k ) says that aeration, catalyzed chemical oxidation, and GAC adsorption will be used in a pilot system for removing organic solvents, including chloroform, carbon tetrachloride, and other low-molecularweight chlorinated hydrocarbons. Pressurized gas aeration followed by direct incineration will be evaluated as an alternative to air stripping, he adds. In this way. discharge of .contaminants from water toair is avoided. A 20-enm plant will test ways to clean wate;ior aquifer recharge at the Rocky Mountain Arsenal, Colo. The GAC portion will use carbon made by Westvaco. Many experts note that there are problems with using GAC for lowmolecular-weight chlorinated hydrocarbon removal from water. Hager says that chemical oxidation using ultraviolet-peroxide systems will address the destruction of nonadsorbable or nonvolatile organic contaminants in the water. Zimpro Inc. (Rothschild, Wis.) is simultaneously using microbes and an organics removal approach involving powdered activated carbon (PAC) with activated sludge. The PAC adsorbs contaminantsthat the microbes cannot assimilate. while the microbes handle organics that otherwise would need an additional carbon step. Zimpro says that its units, presently in use at 15 U S . sites, can remove hydrocarbons, pesticides, and certain other organics from groundwater, wastewater, or leachates, with a 95-99% removal efficiency reported for COD, total nitrogen, and various organics. Known by the trade name “PACT.” the system is being used in a groundwater recharee Droiect in El Paso. Tex., and for&duniwater treatmeni at a chemical plant site in Michigan. But not all contaminants are orEnvirm. ScI. Technol.. VoI. 17. No. 8. 1983
34SA
ganic. Inorganics, such as heavy metals, silica, nitrate, and hardness, are also major groundwater contaminants, against which air stripping, carbon and polymer adsorption, and other approaches are generally of no avail. Removal approaches include pH adjustment, oxidation-reduction, and other adjustments necessary to accomplish insolubilization, filtration, ion exchange, electrodialysis, and reverse osmosis. Hager said that several inorganic treatment processes will be included in the Rocky Mountain Arsenal pilot study to assess heavy metal, silica, and hardness removal efficiencies. At the Nashua, N.H., site, pilot field studies (20 gpm) are being conducted to evaluate metals removal by neutralization and precipitation. Organics (-0.2% by weight) removal and concentration for disposal or destruction through high-temperature air stripping, biological treatment, and distillation also are being studied, according to Bruce Lamarre of Roy F. Weston, Inc. (Concord, N.H.). Design parameters will lead to a 300-gpm treatment facility.
Reclamation problems Both reclamation of contaminated groundwater and aquifer recharge can present technical and legal problems, warns Kenneth Schmidt, a groundwater quality consultant in Fresno, Calif. On the technical side, it often happens that the hydrogeologic system is misunderstood and not all contaminants are identified. These factors, Schmidt says, are especially troublesome for reclamation projects in the Southwest, a region particularly dependent on groundwater. Schmidt spoke of a case involving vineyards south of Fresno, whose sandy soils were treated with dihromochloropropane (DBCP) to kill nematodes. That contaminant’s concentration now exceeds 1 ppb in many wells. The amount of contaminated water involves about 3 million acre-ft near Fresno alone and could involve more than 20 million acre-ft of groundwater in all of California’s San Joaquin Valley. The contamination is usually limited to the upper 200 ft of the aquifer. The problem with defining this contamination is that the source is so diffuse-DBCP was sprayed over many hundreds of thousands of acres of agricultural land from ahout 1955 to 1979. Another difficulty, where contamination plumes are present, is that drawing a contaminant plume into a removal well for treatment can be inefficient. This is because large 350A
Enviran. Sci. Technot.. VOI. 17. NO.8, 1983
An expired law
thorization next year. In any such reauthorization i
there were conclusive proof of from a contaminant. The bell however, that industry‘sprop not be considered. The Senate Toxic Subst Subcommittee has asked the gressional Research Service to pile a bibliography of groundwat formation and to report data ab’ groundwater contamination on state-by-stale basis. amounts of uncontaminated water are often drawn in with contaminated water. Schmidt says that at some reclamation projects, actual contaminated water accounts for less than 20% of the total water pumped. This situation is most likely to occur at sites at which the vertical and horizontal dimensions of the plume have not been precisely delineated before reclamation begins. Schmidt points out that heavy pumping can lead to substantial water table decline and changes in groundwater flow and quality, and in some cases, wells going dry. Also, particularly in the Southwest, contamination plumes can comprise large volumes, sometimes in excess of 100 000 acre-ft. That is a tremendous amount of water to pump and can take a long time, often decades to handle. If large amounts of contaminated water cannot be treated and used, a major disposal problem could result. For example, land treatment by spray irrigation to remove volatile organics can lead to oxidation of ammonia to nitrate, another pollutant that could render the water unfit for use or aquifer recharge. Pumping is not effective in removing contaminants from the vadose zone (the unsaturated zone between the surface and the water table). In wide areas, especially in the Southwest, the vadose zone is commonly several hundred feet thick. Substantial amounts of contaminants can be in this zone and can slowly drain downward to contaminate the “reclaimed” aquifer at a later time. Removal of con-
taminated soils and geologic materials at such depths is not feasible at many sites. As if technical difficulties were not enough, Schmidt calls attention to legal problems. For instance, there are questions of water rights and adverse effects caused by pumping or disposal of treated water. This is especially true in the Southwest where water rights are embodied in a complex set of laws and traditions. Landowners may deny access even for monitoring, let alone groundwater recovery purposes. Siting wells at hydrogeologically preferred locations is often difficult, and the effectivenessof the reclamation program may he limited. Moreover, in some western states, such as Arizona, special permits are necessary for pumping or disposing of contaminated groundwater, and they are not always granted. To compound all of the foregoing problems, there is a shortage of experienced and properly trained groundwater hydrologists, Schmidt observes. In addition, he notes that “unreasonble time constraints for implementing control measures are imposed by some regulatory agencies, often because of intense public pressure. This may cause plans to be implemented before the groundwater system is adequately understood. Serious pollution has resulted from implementing so-called control measures.”
The cost factor A major factor of the feasibility of groundwater reclamation is cost. One cost factor, that of delineating contamination plumes, may be reduced if various geophysical monitoring technologies are refined. Meanwhile, Schmidt points out that hundreds of thousands of dollars must be spent merely to define a plume. Then millions of dollars are required to construct facilities for reclamation. Operating costs are substantial, and many projects will go on for decades before completion. To all this, energy costs must he added. Biological and physical-chemical technologies for treatment (filtration, carbon, reverse osmosis, and others) exist, but they are more effective when the contaminant(s) to be treated are properly identified and quantified and the treatment process does not involve dilution with clean fluid, which reduces treatment efficiency markedly. In the last analysis, the cheapest method of ensuring clean groundwater is, and will continue to be, contamination prevention. -Julian Josephson
